Search

Large scale supersonic bulk flows are present in a wide range of astrophysical objects, from O-star winds to molecular clouds, galactic sheets, accretion, or γ-ray bursts. Associated flow collisions shape observable properties and internal physics alike. Our goal is to shed light on the interplay between large scale aspects of such collision zones and the characteristics of the compressible turbulence they harbor. Our model setup is as simple as can be: 3D hydrodynamical simulations of two head-on colliding, isothermal, and homogeneous flows with identical upstream (subscript u) flow parameters and Mach numbers 2 < Mu < 43. The turbulence in the collision zone is driven by the upstream flows, whose kinetic energy is partly dissipated and spatially modulated by the shocks confining the zone. Numerical results are in line with expectations from self-similarity arguments. The spatial scale of modulation grows with the collision zone. The fraction of energy dissipated at the confining shocks decreases with increasing Mu. The mean density is ρm ≈ 20ρu, independent of Mu. The root mean square Mach number is Mrms ≈ 0.25Mu. Deviations toward weaker turbulence are found as the collision zone thickens and for small Mu. The density probability function is not log-normal. The turbulence is inhomogeneous, weaker in the center of the zone than close to the confining shocks. It is also anisotropic: transverse to the upstream flows Mrms is always subsonic. We argue that uniform, head-on colliding flows generally disfavor turbulence that is at the same time isothermal, supersonic, and isotropic. The anisotropy carries over to other quantities like the density variance - Mach number relation. Line-of-sight effects thus exist. Structure functions differ depending on whether they are computed along a line-of-sight perpendicular or parallel to the upstream flow. Turbulence characteristics generally deviate markedly from those found for uniformly driven, supersonic, isothermal turbulence in 3D periodic box simulations. We suggest that this should be kept in mind when interpreting turbulence characteristics derived from observations. Our simulations show that even a simple model setup results in a richly structured interaction zone. The robustness of our findings toward more realistic setups remains to be tested. [ABSTRACT FROM AUTHOR]

Aims. Observations show nebulae around some massive stars but not around others. If observed, their chemical composition is far from homogeneous. Our goal is to put these observational features into the context of the evolution of massive stars and their circumstellar medium (CSM) and, more generally, to quantify the role of massive stars for the chemical and dynamical evolution of the ISM. Methods. Using the A-MAZE code, we perform 2d-axisymmetric hydrodynamical simulations of the evolution of the CSM, shaped by stellar winds, for a whole grid of massive stellar models from 15 to 120 M and following the stellar evolution from the zero-age main-sequence to the time of supernova explosion. In addition to the usual quantities, we also follow five chemical species: H, He, C, N, and O. Results. We show how various quantities evolve as a function of time: size of the bubble, position of the wind termination shock, chemical composition of the bubble, etc. The chemical composition of the bubble changes considerably compared to the initial composition, particularly during the red-supergiant (RSG) and Wolf-Rayet (WR) phases. In some extreme cases, the inner region of the bubble can be completely depleted in hydrogen and nitrogen, and is mainly composed of carbon, helium, and oxygen. We argue why the bubble typically expands at a lower rate than predicted by self-similarity theory. In particular, the size of the bubble is very sensitive to the density of the ISM, decreasing by a factor of ~2.5 for each additional dex in ISM density. The bubble size also decreases with the metallicity of the central star, because low-metallicity stars have weaker winds. Our models qualitatively fit the observations of WR ejecta nebulae. [ABSTRACT FROM AUTHOR]

We present the parallel particle-in-cell (PIC) code Apar-T and, more importantly, address the fundamental question of the relations between the PIC model, the Vlasov-Maxwell theory, and real plasmas. First, we present four validation tests: spectra from simulations of thermal plasmas, linear growth rates of the relativistic tearing instability and of the filamentation instability, and nonlinear filamentation merging phase. For the filamentation instability we show that the effective growth rates measured on the total energy can differ by more than 50% from the linear cold predictions and from the fastest modes of the simulation. We link these discrepancies to the superparticle number per cell and to the level of field fluctuations. Second, we detail a new method for initial loading of Maxwell-Jüttner particle distributions with relativistic bulk velocity and relativistic temperature, and explain why the traditional method with individual particle boosting fails. The formulation of the relativistic Harris equilibrium is generalized to arbitrary temperature and mass ratios. Both are required for the tearing instability setup. Third, we turn to the key point of this paper and scrutinize the question of what description of (weakly coupled) physical plasmas is obtained by PIC models. These models rely on two building blocks: coarse-graining, i.e., grouping of the order of p ~ 1010 real particles into a single computer superparticle, and field storage on a grid with its subsequent finite superparticle size. We introduce the notion of coarse-graining dependent quantities, i.e., quantities depending on p. They derive from the PIC plasma parameter ΛPIC, which we show to behave as ΛPIC ∝ 1/p. We explore two important implications. One is that PIC collision- and fluctuation-induced thermalization times are expected to scale with the number of superparticles per grid cell, and thus to be a factor p ~ 1010 smaller than in real plasmas, a fact that we confirm with simulations. The other is that the level of electric field fluctuations scales as 1/ΛPIC ∝ p. We provide a corresponding exact expression, taking into account the finite superparticle size. We confirm both expectations with simulations. Fourth, we compare the Vlasov-Maxwell theory, often used for code benchmarking, to the PIC model. The former describes a phase-space fluid with Λ = +∞ and no correlations, while the PIC plasma features a small Λ and a high level of correlations when compared to a real plasma. These differences have to be kept in mind when interpreting and validating PIC results against the Vlasov-Maxwell theory and when modeling real physical plasmas. [ABSTRACT FROM AUTHOR]